Docstring fixes.

sdks/python/apache_beam/ml/rag/embeddings/huggingface.py

@@ -29,20 +29,14 @@
 
 
 class HuggingfaceTextEmbeddings(EmbeddingsManager):
-  """SentenceTransformer embeddings for RAG pipeline.
-    
-    Extends EmbeddingsManager to work with RAG-specific types:
-    - Input: Chunk objects containing text to embed
-    - Output: Chunk objects with embedding property set
-    """
   def __init__(
       self, model_name: str, *, max_seq_length: Optional[int] = None, **kwargs):
-    """Initialize RAG embeddings.
-        
+    """Utilizes huggingface SentenceTransformer embeddings for RAG pipeline.
+
         Args:
             model_name: Name of the sentence-transformers model to use
             max_seq_length: Maximum sequence length for the model
-            **kwargs: Additional arguments passed to parent
+            **kwargs: Additional arguments including ModelHandlers arguments
         """
     super().__init__(type_adapter=create_rag_adapter(), **kwargs)
     self.model_name = model_name

sdks/python/apache_beam/ml/rag/types.py

@@ -34,6 +34,9 @@
 @dataclass
 class Content:
   """Container for embeddable content. Add new types as when as necessary.
+
+    Args:
+        text: Text content to be embedded
     """
   text: Optional[str] = None
 
@@ -42,7 +45,7 @@ class Content:
 class Embedding:
   """Represents vector embeddings.
     
-    Attributes:
+    Args:
         dense_embedding: Dense vector representation
         sparse_embedding: Optional sparse vector representation for hybrid
           search
@@ -56,7 +59,7 @@ class Embedding:
 class Chunk:
   """Represents a chunk of embeddable content with metadata.
     
-    Attributes:
+    Args:
         content: The actual content of the chunk
         id: Unique identifier for the chunk
         index: Index of this chunk within the original document